Is the explanation plausible?

A new "unmapped shortcut" keeps appearing as the best option. Try a few missions with the plausibility guard off, then switch it on and compare.

Upgrades for this level

Capital —

300 cr

Show a calibrated probability next to the raw model score.

500 cr

Show the lower and upper bound around every calibrated probability.

700 cr

See which conditions push the calibrated probability up or down, each with its own uncertainty range.

800 cr

See how the probability and its range would change under other conditions, and carry out the ones that are real actions.

700 cr

Filter explanation candidates whose hypothetical situation is poorly supported by the reference data.